<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
ner-deBERTa-v3-large-conll2003
This model is a fine-tuned version of microsoft/deberta-v3-large on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1546
- Precision: 0.9235
- Recall: 0.9363
- F1: 0.9298
- Accuracy: 0.9853
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0077 | 1.0 | 878 | 0.1280 | 0.9096 | 0.9265 | 0.9180 | 0.9832 |
| 0.0084 | 2.0 | 1756 | 0.1380 | 0.9167 | 0.9299 | 0.9233 | 0.9844 |
| 0.0037 | 3.0 | 2634 | 0.1495 | 0.9221 | 0.9347 | 0.9283 | 0.9850 |
| 0.0015 | 4.0 | 3512 | 0.1517 | 0.9215 | 0.9347 | 0.9280 | 0.9849 |
| 0.0006 | 5.0 | 4390 | 0.1546 | 0.9235 | 0.9363 | 0.9298 | 0.9853 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
